Does Perception Matter in Asset Pricing? Modeling Volatility Jumps Using Twitter-Based Sentiment Indices

Abstract

This article uses public perceptions to forecast short-term fluctuations in asset prices. Based on four billion tweets scraped between 2009 and 2019, I perform textual analysis to construct daily sentiment indices. The sentiment indices allow us to forecast stock volatility jumps as well as expected jump levels. The implications of forecasting volatility jumps are substantive. First, volatility jumps have a significant effect on option prices. Second, changes in the volatility path lead to large (negatively related) changes in the prices’ future trajectory. Determining what information causes jumps allows for better risk management and more accurate asset pricing models.

Keywords:

• Behavioral finance
• Volatility jumps
• News
• Social media
• Big data

JEL Codes:

• G40
• G41
• G14
• G17

Disclosure statement

I wish to confirm that there are no known conflicts of interest or competing interests associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.

Table 10. Markov-switching regressions: Apple tweets June 2009–February 2019.

Notes

1 But please note that Manela and Moreira (2017) have found that news can in fact predict above average stock returns and periods of economic disasters.

2 And, by extension, for risk management purposes.

3 I would argue that the majority of tweets made by the general public (not investment professionals) with respect to a specific company are based on spur of the moment feelings, not firm fundamentals.

4 What is meant by noise is a stock’s deviations from the fundamentally valued price.

5 Throughout this article, the terms “short-term fluctuations” and “noise” are used interchangeably.

6 This was calculated by looking at the most generous data download service from Twitter for non-enterprise downloading: the Premier tier. At this tier, Twitter allows users to make 2,500 download requests per month and each request is capped at 500 tweets at a cost of $1,899.99 per month. This information is available directly from Twitter’s website here: https://developer.twitter.com/en/pricing.html.

7 This is done by other programs that conduct textual analysis such as the StockTwits.